Error-Related EEG Potentials Generated During Simulated Brain–Computer Interaction

TLDR

Brain‑computer interfaces are error‑prone, and error‑related potentials (ErrP) recorded in EEG after an error can be used to verify and improve BCI accuracy; ErrP have been observed in typical choice‑reaction tasks. The study investigates whether ErrP are elicited when the interface itself makes an error during intent recognition, and confirms the presence of a new kind of ErrP in a human‑robot interaction experiment with five healthy volunteers. A classifier trained on data recorded up to three months earlier achieved 83.5 % accuracy for correct trials and 79.2 % for erroneous trials. To exploit these ErrP, they must be detected in each single trial using a short window following the BCI feedback. The authors explored whether ErrP follow feedback indicating incorrect responses of the simulated BCI interface.

Abstract

Brain-computer interfaces (BCIs) are prone to errors in the recognition of subject's intent. An elegant approach to improve the accuracy of BCIs consists in a verification procedure directly based on the presence of error-related potentials (ErrP) in the electroencephalogram (EEG) recorded right after the occurrence of an error. Several studies show the presence of ErrP in typical choice reaction tasks. However, in the context of a BCI, the central question is: "Are ErrP also elicited when the error is made by the interface during the recognition of the subject's intent? "; We have thus explored whether ErrP also follow a feedback indicating incorrect responses of the simulated BCI interface. Five healthy volunteer subjects participated in a new human-robot interaction experiment, which seem to confirm the previously reported presence of a new kind of ErrP. However, in order to exploit these ErrP, we need to detect them in each single trial using a short window following the feedback associated to the response of the BCI. We have achieved an average recognition rate of correct and erroneous single trials of 83.5% and 79.2%, respectively, using a classifier built with data recorded up to three months earlier.